0
$\begingroup$

Question: $(x_n)_{n=1}^\infty$ is a sequence with $x_n\neq0 $ for all $n$, also let $x_n$ tend to infinity. Let $(y_n)_{n=1}^\infty$ be defined by $y_n=\frac{1}{x_n}$, show that this converges to zero.

Definition for tending to infinity:$\forall K \in \mathbb{R} \exists N\in \mathbb{N} \forall n \in \mathbb{N} ,n>N:x_n>K $

Definition for convergence to zero:$\forall \varepsilon > 0 \exists N\in \mathbb{N} \forall n \in \mathbb{N} ,n>N:∣x_n∣<\varepsilon $

Idea: I know that $x_n$ tends to infinity since this is assumed. Then it must satisfy the defintion, therefore there must exist an $N$ which satisfies the defintion which i will call $N_x$ to prevent confusion.

To prove that $y_n$ converges to zero then I should choose $N_y=\lceil\frac{1}{N_x}\rceil$

Proof (attempt): Given $\varepsilon>0$ choose $N_y=\lceil\frac{1}{N_x}\rceil$. Given $n \in \mathbb{N}$, $n>N$ we have $∣y_n∣$$=∣\frac{1}{x_n}∣\leq \frac{1}{N_x}\leq\varepsilon $

Not entirely sure my selection for $N_y$ is correct, I understand why the statement is true however the proving put I am struggling.

$\endgroup$
  • $\begingroup$ For any $\varepsilon >0$ there exists $M \in \Bbb{N}$ such that $1/M< \varepsilon$ solves this, as $1/n$ is obviously decreasing $\endgroup$ – B.Swan May 11 at 20:53
2
$\begingroup$

Take $\varepsilon>0$. Since $\lim_{n\to\infty}x_n=\infty$, there is some $N\in\mathbb N$ such that $n\geqslant N\implies x_n>\frac1\varepsilon$. But then $n\geqslant N\implies 0<\frac1{x_n}<\varepsilon$. In particular, $\left\lvert\frac1{x_n}\right\rvert<\varepsilon$.

$\endgroup$
  • $\begingroup$ Hi thank you for your reply, would you mind clarifying how it implies that $x_n>\frac{1}{\varepsilon}$? $\endgroup$ – ViB May 11 at 21:17
  • $\begingroup$ I am using the definition of $\lim_{n\to\infty}x_n=\infty$, taking $K=\frac1\varepsilon$. Is it clear now? $\endgroup$ – José Carlos Santos May 11 at 21:21
0
$\begingroup$

Your intuition is right: the definition of $N_y$ is a bit off.

When you choose $N_x$, this value is for a particular $K$, from the "For all $K$ there exists $N\in\mathbb{N}$" part of the first statement. As $K$ increases without bound, so will $N_x$. Here, I think you're choosing $N_y$ with the goal of showing that $y_n$ is sufficiently small for all $n\geq N_y$. But if you let $K\rightarrow\infty$, choose a corresponding $N_x$, and then let $N_y=\lceil\tfrac{1}{N_x}\rceil$, this will result in $N_x$ shooting off to $\infty$ but $N_y$ will end up being $1$ all the time. Your goal will then be proving that $y_n$ is sufficiently small for all $n\geq N_y=1$, which won't be true.

For this problem, you've been given a relationship between the values of $\{x_n\}$ and $\{y_n\}$, so you should use this relationship to somehow connect the formal statements of "$x_n\rightarrow\infty$" and "$y_n\rightarrow 0$". Let's restate your proof goal:

Choose any $\epsilon>0$ (this choice is out of your control, your proof must work for any choice). You want to produce an $N\in\mathbb{N}$ such that $\{y_n\}$ satisfies the definition.

First, pick some $K>0$ (you're allowed to do this since we know that $x_n\rightarrow 0$) such that if $x_n>K$ then $y_n<\epsilon$. This $K$ will be expressed algebraically in terms of $\epsilon$. (EDIT: I think you responded to another post very recently with this choice of $K$ - use that one.) Then, if you take some $N_x$ such that $x_n>K$ for all $n\geq N_x$, then you'll be able to say something about $N_x$ and $\{y_n\}$ similarly.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.